Breast Cancer Prediction Using Ensemble Techniques Manikandan G.1,*, Karthikeyan B.1, Rajendiran P.2, Harish R.3, Prathyusha T.3, Sethu V.3 1Senior Assistant Professor, School of Computing, SASTRA Deemed University, Thanjavur, India 2Assistant Professor, School of Computing, SASTRA Deemed University, Thanjavur, India 3Student, School of Computing, SASTRA Deemed University, Thanjavur, India *Corresponding Author: G. Manikandan, Senior Assistant Professor, School of Computing, SASTRA Deemed University, Thanjavur, India, Email: manikandan@it.sastra.edu
Online published on 19 August, 2019. Abstract One of the most dangerous types of cancer affecting women across the world happens to be breast cancer. As per clinical experts, detecting this cancer in its first stage is crucial in saving lives. The variables like sample code number, clump thickness, uniformity of cell size, uniformity of cell shape etc., are some of the important risk factors providing information, allowing identification of the recurrence of breast cancer and cure it in the earliest stages. The factors responsible are obtained as an output from the data mining models. Existing studies and their results have produced good accuracies with minimal error rates in prediction. The medical diagnostic models which are based on data mining techniques can capture delicate designs and dependencies providing promising results. Promising results can be obtained from medically focussed diagnostic models which are based on data mining techniques, which have the capability to interpret ethereal structure and relationships. Single data mining technique may not suffice in providing results which are constant and competent. Ensemble learning involves the blending of combination rules with input as several selected and generated intermediate risk factors to finally obtain simple selection results as output. The aim of this paper is to analyse the existing standard methodologies and to develop a new ensemble model comprising of classification techniques which will result in higher precision. Top Keywords Breast Cancer, Prediction, Voting, Stacking, Boosting. Top |